THE CHEMICAL ELEMENTS AND THEIR COMPOUNDS

BY*

N. V. SIDGWICK FELLOW OF LINCOLN COLLEGE

HON. STUDENT OF CHRIST OHUBCH FORMERLY PROFESSOR OF CHEMISTRY

IN THE UNIVERSITY OF OXFORD

VOLUME I

O X F O R D AT THE CLARENDON PRESS

Oxford University Press, Amen House, London E.C. 4 OhABQOW NBJW YOBK TOBONTO MELBOUBNEI WELLINGTON

BOMBAY CALCUTTA MADBAS OAPB TOWN

Geoffrey Cumberlege, Publisher to the University

IIAlV f triLIIXID 1060 Heprlntad lithographioally Ia Groat Brltntn

Ht lhi VMliailf Y PMIl1 SMfOlD, 10Sl

PREFACE

THIS book is an attempt to discuss in detail the properties of the elements and their compounds in the light of modern ideas of atomic and molecular structure. The development of these ideas in the last thirty years has made it possible to transform 'Inorganic5 Chemistry from a mass of disconnected facts into an ordered system of relations. Inorganic text-books are, however, usually so over-burdened with the details of mineralogy, metallurgy, technical chemistry, and analysis that hardly any space is left for the considerations of the theoretical relations; while the customary exclusion of all but the simplest compounds of carbon deprives us of the help of the best-known and most important of the elements. I have tried to avoid these errors both of excess and of defect, and to give an account of the compounds of all the elements, with special reference to the general relations between them. The basis of classification is, of course, the Periodic Table, the simplest arrangement for exhibiting the relations of the elements.

Among my many obligations I must first express my indebtedness to the great Handbooks of Abegg and of Gmelin (8th edition), as well as to the Abstracts published by the British and the American Chemical Societies: to Professor Linus Pauling's Chemical Bond (Cornell University Press, 2nd edition, 1942): and to the Structural Inorganic Chemistry of A. F. Wells (Clarendon Press, 1945), The numerous valuable monographs on particular branches of the subject that I have consulted are acknowledged in their places. But I would particularly mention the Thermochemistry of F. R. Bichowski and F. D. Rossini (Reinhold, New York, 1936): G. W. Wheland's Theory of Resonance and its Application to Organic Chemistry (Wiley, New York, 1944): and the Chemie der metall-organischen Verbindungen of E. Krause and A. v. Grosse (Borntraeger, Berlin, 1937).

Among those from whom I have received personal help I must in the first place express my deep gratitude to Professor Linus Pauling, who, in addition to publishing a book of the first importance on molecular struc-ture, has, especially during his residence in Oxford as Eastman Professor, given up much time to answering my questions on a whole series of points large and small, and has read and criticized in detail my introductory section. To my colleagues in Oxford, and especially to Dr. L. E. Sutton of Magdalen College, and to Mr. H. M. Powell, University Reader in Chemical Crystallography, I am very grateful for help and advice on many points. I owe more than I can say to the late Dr. R. V. G. Ewens, formerly Scholar of this College, and Reader in Chemistry at Guy's Hospital Medical School, who up to his untimely death had read nearly the whole of my manuscript, and made numerous corrections and modifications of the greatest value to me. Dr. M. W. Lister of Harwell, now Assistant Professor of Chemistry at Toronto University, has read the whole book in proof and detected many rrori, I have bean greatly helped by him and by Dr. Gharlei Coryell of the

vi PREFACE Massachusetts Institute of Technology in my account of the uranide elements. In the laborious checking of the numerous references I have had t,Jio assistance of Mr. W. T. L. Neal of Exeter College, Mr. B. B. Goalby of Now College, Mr. M. F. Hoare of The Queen's College, and Mr. A. Mackay of this College. The indexes were compiled with the help of Mr. C. P. JH organ of Trinity College.

The references I have tried to bring up to the summer of 1948; but the dolays in the publication of original papers and abstracts, and in some wubjacts the restrictions of military secrecy, make it peculiarly difficult to ttttwign an exact term to this in the immediate post-war years. Many of the Ntatements in this book already need correction and amplification, but that is inevitable in so rapidly advancing a subject. LINCOLN COLLEGE

OXFORD May 1949

CONTENTS VOLUME I

Abbreviations . . . . xi Some Useful Physical Constants . . . xii Introduction . . . . . . . . xv

Fundamental Particles (xv). Atomic Structures (xvi). Periodic Classification (xvi). Molecular Structure (xvii). Resonance (xvii). Electronegativity (xviii). Stereochemistry (xix). Atomic Radii (xxi). Heats of Linkage (xxii).

Tables . . . . . . . . xxiii I : International Atomic Weights (xxiii). I I : Atomic Structures (xxiv). I l l : The Periodic Classification (Thomsen-Bohr) (xxvii); IV": (MendeleerT) (xxviii). V: Atomic and Ionic Radii (xxix). VI: Heats of Atomization (Hf) from Standard States (xxx); (Ha) of Links from Atoms (xxxi).

Group 0 # . . . . . . . 1

INEBT GASES, General (1): Physical Properties (1), HELIUM (1). Liquid and Solid Helium (5). OTHER INEBT GASES (8): Chemical (?) Compounds (8).

Group I . . . . . . . . 11 HYDROGEN (11): Physical Properties (12). Monatomic Hydrogen (15). The ProtonH+(18). Binary Hydrides (21). The Hydrogen Bond (23). DEUTERIUM (33). Properties of its Compounds (36). Heavy Water (44). Exchange of Hydrogen and Deuterium (50). TRITIUM, Preparation and Properties (57).

Group I A. Alkali Metals . . . . . . 59 Hydrides (66). Derivatives of Hydrocarbons (67). Nitrides (84). Oxides (89). Halides (93). Salts (95). Complexes (97). Peculiarities of LITHIUM (100).

Group I B. Copper, Silver, Gold . . . . . 1 0 3 Univalent Compounds (106). Oxides (117). Halides (119). Salts (125). Univalent Complexes (129).

Polyvalent Compounds (148). CUPRIO Compounds (148). Com-plexes (156). ARGENTIC Compounds (174). Complexes (176). AURIC Compounds (177). Alkyls and Aryls (181). Complexes (188).

Group II . . . . . . . 1 9 3 BEKYLLIUM (197). Compounds (199). Salts (205). Complexes (208).

Group II A. Alkaline Earths . . . . . 2 1 9 Solubilities and Water of Crystallization of Salts of the Elements Beryllium to Radium (220). MAGNESIUM (222), Alkyls and Aryls: Grig-nard Reagents (224). Salts (234), Complexes (241). ALKALINE EABTMS(242). Compound! (246). Salt* (251). Complexes (2SB).

viii CONTENTS Group II B. Zinc, Cadmium, Mercury . . . . 262

Zinc and Cadmium (263). Compounds (264). Salts (271). Complexes (277).

M ETtCUBY (285). Mercurous Ion and Compounds (289). Complexes (295). MBRCUBIC Compounds (296). Carbon Compounds (298). Other Mercuric Compounds (317). Complexes (327).

Group III . . . . . . . . 334 HoitoN (336). Boron Hydrides (338). Derivatives of Boranes (348). Horohydrides (364). Normal Boron Compounds (367). Complexes (308)/

ALUMINIUM (412). Alkyls and Aryls (414). Compounds (418). Complexes (429).

Group III A. Scandium, Rare Earth Elements, Actinium . 439 HcAN DiUM (440). YTTRIUM (441). The LANTHANIDES (441). Physical Peculiarities (442). Compounds (446). Abnormal Valencies (450). ACTINIUM (457).

Group III B. Gallium, Indium, Thallium . . . . 4 5 8 Trivalcnt States. Alkyls and Aryls (460). Salts (466). Complexes (472). Divalent Gallium and Indium (477). Monovalent Indium (480). Monovalent ThaUium (481). Complexes (487).

Group IV. Carbon . . . . . . . 488 CARBON (490). Organic Compounds, saturated (496): unsaturated (505). Theories of Organic Reactions (513). Oxides of Carbon (521). Organic Oxygen Compounds (523). Trivalent Carbon (529). Divalent Carbon (545). Carbonyl Compounds (547).

Group IV B. Silicon, Germanium, Tin, Lead . . . 5 5 1 Tetravalent Compounds. Hydrides (553). Organic Compounds, Sum-mary (555). Silico-organic Compounds (559). Germanium Organic Compounds (568). Tin Organic Compounds (578). Lead Organic Compounds (587). Nitrides (598). Oxides (599). Sulphides (603). Totrahalides (605). Complexes (612): with Nitrogen (612): Oxygen (613): Halides (614).

Divalent Elements (616): Silicon (617): Germanium (617): Tin (619): Lead (623).

Group IV A. Titanium, Zirconium,,Hafnium, Thorium . . 628 Metals (631). Hydrides, Carbides (633). Nitrides, Oxides (634). Halides (637), Salts (639). Complexes (640). HAFNIUM (646). Lower Valencies: Titanium (649): Zirconium (651).

Group V. Nitrogen . . . . . . . 654 NiTBoaiN (656). Ammonia and Derivatives (658). Binary Nitrides (603). Organic* Nitrogen Compounds (604). Cyanogen Compounds (007). Oxidei of Nitrogen (081). Oxy-aoids of Nitrogen (003), Nitro-gen Halldos (704). N - N Compound* (703). Free Nitrogen Radicals (718).

CONTENTS IX Phosphorus . . . . . . . . 725

Hydrides (728). Nitrides (733). Oxides (736). Oxy-acids (739). Halides (752). Complexes (755).

Group V B. Arsenic, Antimony, B i s m u t h . . . . 758 Organic Compounds of Arsenic (761): of Antimony (772): of Bis-muth (779). Oxides and Oxy-acids (784). Halides (791). Complexes (797).

Group V A. Vanadium, Niobium, Tantalum, Protoactinium . 804 VANADIUM Compounds (805). Complexes (817). Vanadium, tetra-valent (818): Trivalent (823): Divalent (832). NIOBIUM (834). Penta-valent Complexes (839): Lower Valencies (841). TANTALUM (843): Pentavalent (845). Lower Valencies (849). PROTOACTINIUM (851).

VOLUME I I Group VI . . . . . . . . 855

OXYGEN (856). Ozone (859). Water (863), Hydrogen Peroxide and its Derivatives (868).

SULPHUR (875). Hydrogen Sulphides (878). Organio Sulphides (880). Nitrogen Sulphides (892). Oxides of Sulphur (894). Oxy-acids of Sulphur (904). Organic Derivatives of Oxides and Oxy-acids (921). Oxy-halides (928). Thionic Acids (940). Sulphur Halides (943).

SELENIUM, TELLURIUM, POLONIUM (948). Hydrides (951). Organio Derivatives of Selenium (953): of Tellurium (964). Oxides of Selenium (970): Oxy-acids of Selenium (972). Oxides and Oxy-acids of Tel-lurium (980). Selenium Halides (986). Tellurium Halides (990).

POLONIUM (995). Group VI A, Chromium, Molybdenum, Tungsten, Uranium and

the Uranides . . . . . . . 9 9 8 CHROMIUM (1000). Hexavalent Chromium (1003). Penta- and Tetra-valent Chromium (1008). Trivalent Chromium (1009). Chromic Com-plexes (1014). Divalent Chromium (1022). Chromium Carbonyl (1026). MOLYBDENUM and TUNGSTEN (1028). Hexavalent Compounds (1032). Halides (1033). Oxides (1037). Oxy-acids (1038). Pentavalent Compounds (1047). Tetravalent (1053). Lower Valencies (1057). Carbonyls (1066).

URANIUM (1069). Hexavalent (1071): Tetravalent (1080): Tri-and Divalent (1085). Nuclear Fission (1087). Properties of the 'Uranide* Elements (1091). Chemistry of NEPTUNIUM (1093), PLUTONIUM (1094), AMERICIUM (1095), and CURIUM (1096).

Group VII B. THE HALOGENS . . . . . 1097 FLUORINE (1099). Hydrogen Fluoride (1102). Inorganic Fluorides (1112). Organic Fluorine Compounds (1116). Oxides of Fluorine (1135).

CHLORINE,BBOMINE,IODINE(1139). Inter-halogen Compounds (1146). Hydrogen Halides (1160). Binary Halides (1170). Organio Com-pounds of Chlorine, Bromine, and Iodine (1174). Formation (1174);

X CONTENTS Physical and Chemical Properties (1184). Perhalides M[XJ (1190). Halogen Oxides (1201). Oxy-acids (1212). Compounds of Polyvalent Halogens (1243).

ASTATINE, NO. 85 (1260). Group V I I A . . . . . . . . 1262

MANGANESE (1264): Heptavalent (1266): Hexavalent (1269): Tetra-valont (1271): Trivalent (1274): Divalent (1282). Nitrosyl Com-pounds (1288). TECHNETIUM, No. 43 (1289). RHENIUM (1291): Hepta-valent (1294): Hexavalent (1300): Pentavalent (1303): Tetravalent (1306). Lower Valencies (1311). Carbonyls (1314).

Group VIII. Iron, Cobalt, Nickel . . . . . 1316 IRON (1319). Ferrous Compounds (1327): Complexes (1335). Ferric Compounds (1348): Complexes (1358). Carbonyls (1369). Mtrosyls (1372).

COBALT (1375). Divalent (1376): Trivalent (1392). Cobaltie Com-plexes (1396). Tetravalent Cobalt (1420). Carbonyls (1422). Nitro-wyls (1423).

NICKEL (1426). Non-valent and Monovalent (1429). Divalent (1430). Divalent Complexes (1438). Trivalent Nickel (1449). Car-bonyls (1451). Mtrosyls (1452).

The Platinum Metals . . . . . . 1454 Group VIII A. Ruthenium and Osmium . . . . 1455

Valencies (1455). RUTHENIUM (1459): Divalent (1460): Trivalent (1465): Tetravalent (1475): Penta- (1478), Hexa- (1479), and Hepta-valent (1480): Octovalent (1481). Carbonyls (1482). Nitrosyls (1484).

OSMIUM (1490). Divalent (1490): Trivalent (1492): Tetravalent (1493): Hexavalent (1499): Octovalent (1503). Carbonyls (1509). Nitrosyls (1510).

Group VIII B . Rhodium and Ir idium . . . . 1 5 1 1 RHODIUM; (1513). Divalent (1515): Trivalent (1516). Complexes (1520). Higher Valencies (1527). Carbonyls (1528).

IRIDIUM (1530). Divalent (1531): Trivalent (1532). Complexes (1535). Tetravalent (1541). Complexes (1544). Hexavalent (1546). Carbonyls (1548).

Group VIII C. Palladium and Plat inum . . . . 1550 PALLADIUM (1553). Divalent (1558). Complexes (1561). Trivalent (1573): Tetravalent (1574). Complexes (1575). Carbonyls (157V). Nitrosyls (1577).

PLATINUM (1578). Divalent (1581). Complexes (1583). Trivalent (1005): Tetravalent (1611). Complexes (1615). Hexavalent (1625). Carbonyls (1627). Nitrosyls (1629)*

Author Index . . . . . . . 1629 Subject Index . . . . . . . 1670

ABBREVIATIONS Journals

Ann. Liebig's Annalen der Chemie. Ann. Rep. Chem. Soc. Annual Reports of the Chemical Society. Atti R. AtH delta Eeale Accademia dei Lincei, Ber. Berichte der Deutschen Ghemischen Oesellschaft. Bull. Soc. Bulletin de la Sociiti Chimique de France. C or Chem. Centr. Chemisches Centralblatt. C.E.N. Chemical and Engineering News (Washington). CR. Gomptes rendus hebdomadaires des Sdances de VAcaMmie des Sciences. Gaz. Gazzetta Chimica Italiana. J.A.C.S. Journal of the American Chemical Society, J.CS. Journal of the Chemical Society. J.S.CI. Journal of the Society of Chemical Industry. Mon. Monatshefte fur Chemie. Trans. Far. Soc. Transactions of the Faraday Society.

Books Abegg. R. Abegg et al. Handbuch der anorganischen Chemie. Hirzel, Leipzig,

1905- . Gmelin. Qmelin's Handbuch der anorganischen Chemie. Verlag Chemie, Berlin,

8th edition, 1924- . Chem. Bond. L. Pauling, The Nature of the Chemical Bond. Cornell Press, 1939,

2nd edition, 1942. E.T.V. N. V. Sidgwick, Electronic Theory of Valency. Clarendon Press, 1927. Krause and v. Grosse. E. Krause and A. v. Grosse, Die Chemie der metall-

organischen Verbindungen. Borntraeger, Berlin, 1837. M.I. C. Willgerodt, Organische Verbindungen mit mehrwertigem Iod, Enke,

Stuttgart, 1914. N. Ansch. A. Werner, Neuere Anschauungen auf dem Gebiete der anorganischen

Chemie. Vieweg, Brunswick (1st edition, 1905; 3rd, 1913; 5th, revised by P. Pfeiffer, 1923).

NN. I. and W. Noddack, Das Rhenium. Voss, Leipzig, 1933* S.I.C A. F. Wells, Structural Inorganic Chemistry. Clarendon Press, 1945. Wheland, Resonance. Q. W. Wheland, The Theory of Resonance and its Applica-

tion to Organic Chemistry. New York and London, 1944. c -Veloci ty of light |

s e e p . x i L h = Planck s Constant' v = Frequency (vibrations per second). E.AiN. Effective Atomic Number (see p. xvii). Polar should be used only in the strict sense, meaning 'having a dipole moment*. SoPi grg, anhydrous substance dissolving in 100 grs. solvent.

3Ql ABBREVIATIONS Square brackets [ ] : (1) in reaction kinetics mean concentrations (usually in

moles per litre); (2) in formulae indicate ions. et al. = and others (authors).

Greek and Latin numerical prefixes (uni = mono; ter = tri; sexa = hexa, &c.) are used indifferently; the supposed objection to 'hybrid' words, of which the first part is derived from Greek and the second from Latin, if it were valid would require us to say quadrifluoride but tetrachloride, would reject 'metastable', and would condemn the Church of England for speaking of the Pananglican Synod. Note that ennea = 9 and dodeca = 12. Symbols like al, hg, mean atom/valency (al = J Al : hg = | Hg).

In types of formulae A is as a rule put for the central atom and B for the covalently attached atoms. E is used for alkyls or aryls, X for halogens and other monovalent radicals (ionized or not), and Am for NH3 and similar amines. Other abbreviations are AIk = alkyl; Ar = aryl; Me, Et, Pr, Bu, &c. = methyl, ethyl, propyl, butyl, &c.; = phenyl (O6H5); Bz = benzyl (C6H6 'CH2); py = pyridine; en orettethylenediamine; Ox = oxalato-group (C2O4); Cy = CN. A is sometimes used for the monovalent radical of a diketone or keto-ester (as C6H7O2 from C6H8O2, apetyl acetone).

SOME USEFUL PHYSICAL CONSTANTS Lengths : 11* = 10~*; 1 fifi = 10~7; 1 A or A.U. = 10~8 cm. 1 X unit for X-rays

= 1/1000 A (strictly 1/1002 A). Velocity of light, c = 2-99776 X1010 cm./sec. Quantum Theory: Energy E =* hv, where v = frequency (vibrations/second) and

h (Planck's constant) = 6-620 xl0~ 2 7 erg-seconds. Absolute zero = 0 K. = -273-16 C. 1 calorie = 4-183 X107 ergs = 4-183 joules. 1 electron-volt (e.v., energy acquired by an electron in moving through a

potential drop of 1 volt) is equal to 1-691X 10~12 erg, or 23-07 k.cals. per g.-molecule.

1 Faraday = 96,500 coulombs.

Light Quanta Wave-number = waves per cm. = frequency/c. Quantum of wave-length 7,000 A (red end of visible) = 40-8 k.cals./g.-mol.: of

wave-length 4,000 A (violet end) =71-4 k.cals./g.-mol.

Factors for Absorption and Emission Spectra

Transition Electronic

Oscillational or Vibrational

Rotational , , ,

Energy

E. volts 1-10

1/10

1/1000

Htjg.-mol. 23-230 k.cals.

2*3 k.oale.

S3 calf,

Wave-length 1^,350 to 1,235 A

123,500 A - 12-85 ja

1,888 ^ m 0*1180 em.

Wave number.

cm."*1

8,100 tq 81,000

810

8 1

PHYSICAL CONSTANTS XHl Einstein's Equivalence of Mass and Energy

E = mc2. Hence the mass of one hydrogen atom corresponds to 940 million e.v., or of 1 gramme to 2-2 X1010 k.cals.

Gas constant B = 1-9885 cals./degree C. Avogadro Number N (molecules in 1 g.-molecule) = 6-025 XlO23. Mean translational energy of N gas molecules at T0 is BT9 = ca. 600 k.cals.

at 25 C. The fraction NJN of the molecules of a gas that have energies of nE or above

(where E is the mean energy) is for W= 5 10 50

NJN 6-8 x 10-8 4-5 X 10-5 3-7 x 10~22

There are two scales of atomic weights in use, one physical and the other chemical. On the physical scale the unit is ^ t h of the mass of the commonest isotope of oxygen (16O = 16*000); on the chemical scale it is ^ t h of the mean atomic mass of the ordinary mixture of oxygen isotopes (0 = 16-000). Accord-ing to Aston (1942) the ratio

Unit of Chemical Atomic Weight __ - . / ^ 9 7 * Unit of Physical Atomic Weight ~~

INTRODUCTION FUNDAMENTAL PARTICLES1'2

THE fundamental particles of which the universe is now believed to be composed are given in the following table, with their dates of discovery, electric charges (e = 4-802 x 10~*10 E.S.U.),3 masses (on the physical scale, where 16O = 16*000), and their average lives.

Particle Proton Neutron Electron

Positron Meson*

Neutrino

Date of discovery

. .

1932 1897

1932 1935 1934

Charge

+ e 0

e

+e fe, +e \ (?a lso0)

0

Mass

1-00758 l-00894a'& 0-000548c (H/1838) 0-000548

ca. H/9) and H/6/

? 10~18

Life Infinite Infinite Infinite

ca. 10~9 sec. ca. 10~6 sec.

a = , b = 6, c = .

Of these the first three alone are of primary importance for chemistry. The proton is, of course, the hydrogen nucleus or positive hydrogen ion (see I. 26).

The neutron9*10 with a mass number of 1 and an atomic number of zero may be called the first element of the Periodic Table, but as it can hold no electrons it has no chemical properties. Owing to the absence of charge it has an enormous penetrating power; while a proton of velocity 30,000 km./sec. (c/10) will travel only one foot in air, a neutron may go several miles in air before it loses all its energy, making only a few collisions on the way. The absence of charge also makes the neutron a very effective pro-jectile for nuclear disruption, since it is not repelled by the nuclear charge as an a-particle would be. If neutrons could be concentrated they would

* There probably7*8 are at least two kinds of mesons, one with a mass equal to about 200 electrons, and the other about 320. The heavier kind were made arti-ficially in the big Berkeley cyclotron, by passing 380 m.e.v. a-particles through thin plates of beryllium, carbon, or copper.8

1 R. E. Peierls, Nature, 1946, 158, 773.

2 L. Pauling, General Chemistry, S. Francisco, 1947, p. 570.

3 V. D. Hopper and T. H. Laby, Proc. Boy. Soc. 1941, 178, 243; T. H. Laby,

Nature, 1942, 150, 648. 4 D. J . Hughes, Phys. Rev. 1946, ii. 70, 219. 5 W. E. Stephens, Rev. Mod. Phys. 1947, 19, 19.

6 R. T. Birge, Phys. Rev. 1941, ii. 60, 766.

7 J . Ruling and R. Steinmauer, Experientia, 1946, 2, 108.

8 See Chem. and Eng. News, 22 Mar. 1948 (p. 850).

9 J . Chadwick, Nature, 1932, 129, 312; Proc. Roy. Soc. 1932, 136, 692; ib. 1933,

142, 1 (Bakerian Lecture). 10

See also P . B, Moon, Ann. Rep, Chem. Soc. for 1938,35, 21.

xvi INTRODUCTION form a gas half as dense as hydrogen; the idea that liquid neutrons would have an enormous density (with a radius of 2x 10~13 cm. one c,c. would weigh 25 million tons) is fallacious; the half quantum of zero-point energy (see under helium p. 7) would bring its effective radius nearly up to that of an ordinary atom.

A list of the International Atomic Weights of the elements is given in Table I (p. xxiii).

ATOMIC STRUCTURES The nucleus of every atom except hydrogen is made up of protons and

tieutrons; if the mass number is N and the atomic number Z it consists of X protons and JVZ neutrons; two isotopes have the same Z but different ATs. In a /?-ray change we must suppose that a neutron changes into a proton with the emission of an electron. In the neutral atom the nucleus m surrounded by as many electrons as it contains protons; the atomic number is (1) the ordinal number of the element in the periodic system, (2) the number of protons in, and hence the positive charge of, its nucleus, and (3) the number of electrons surrounding the nucleus in the neutral isolated atom.

The electrons are arranged* in groups or shells according to their prin-cipal quantum numbers 1, 2, 3, &c. (K, L, M, N, O, P, Q , . . . ) ; the elec-trons of each group are further divided into subgroups (s, p,d,f,...); the maximum number* of subgroups is equal to the principal quantum number, and the largest number of electrons that each subgroup can contain is:

Subgroup . . 8 p d f Max. No. . . 2 6 10 14

Hence the maximum size of the groups of principal quantum numbers 1, 2, 3, 4, . . . n is 2, 8, 18, 32, . . . 2ft2.

A list of the structures of the elements is given in Table II, p. xxiv.

THE PERIODIC CLASSIFICATION The periodic relations of the elements (Newlands, 1864; Mendeleeff,

1889; Lothar Meyer, 1870) can be expressed in two ways, each of which has its advantages. The form adopted by Bohr (Table III), in which each period, beginning and ending with an inert gas, is written in one line, shows most clearly the development of the atomic structures. The elements in brackets are those with an incomplete (between 8 and 18) electronic group In the core (i.e. as well as the outermost valency group); those within double brackets (the lanthanide and uranide elements) have two such groups, the second being between 18 and 32.

The seoond form of the table, due originally to Mendeleeff (Table IV), is more useful for bringing out the chemical similarities (which arc so

* For an explanation of tilt atomio stmefcurss s L. Pauling and 1. Bright WUsOa1 Jr,, Introduction to Quantum M*ohcmio*t 3HoOrHw-HiIl, Now York, 1085.

INTBODUCTION XVIi dependent on the valencies), and is the one adopted in the following chapters. There are nine groups, each except the first and the last with two subgroups. The elements of the JVth group are (with an obvious modification for Group VIII and for the lanthanides and the uranides) those which have either N electrons more than the preceding, or S-N less than the following inert gas; in the first two (typical and sub-typical) periods one element satisfies both conditions; in the later periods there are two, the elements of the first kind forming subgroup A, and those of the second subgroup B.

MOLECULAR STRUCTURE The atoms in a molecule are held together through their electrons, and

essentially in two ways: (1) by the transfer of electrons from one atom to another, giving rise to an electrostatic (Coulomb) attraction (electro-valency, Kossel, 1916), and (2) by the sharing of pairs of electrons between two atoms, so that in a sense they belong to both (covalency: G. N. Lewis, 1916); the pair may either come one from each atom (normal) or both from one of them (co-ordinate or dative). The conditions which favour the passage of an electrovalency into a covalency are (Fajans, 1923-5) (1) a large charge on the ion, (2) a small size of the cation, (3) a large size of the anion, (4) the possession by the cation of a structure which is not that of an inert gas.

The effective atomic number (E.A.N.) of an atom in a molecule is the number of electrons which it has after these changes, and so is the atomic number plus 1 for each anionic charge, and for every electron from another atom which it shares, and minus 1 for each electron which it loses in becom-ing a cation. The valency of the atom is the difference between the number of unshared electrons in the isolated atom (the atomic number) and the number in the combined atom (Grimm and Sommerfeld, 1926); its covalency is the number of pairs of shared electrons that it holds. The maximum value of the covalency is limited (save under exceptional circumstances) in accordance with the period of the atom in the table, being 4 for the first short period (LiF), 6 for the second short and first long periods (Na-Cl, KBr), and 8 for the heavier elements.

RESONANCE11 This crude picture of atomic linkages is considerably modified in prac-

tice. Covalent links can sometimes be formed by one or by three electrons, and the links in molecules are often (perhaps usually) of an intermediate or mixed character, owing to the phenomenon of resonance. The equations of wave mechanics show that if a molecule can be represented, on the ordinary structural theory, by two different structures, then under certain

11 See Pauling, especially Chemical Bond, ed. 2, pp. 124-59; G. W. Wheland,

Theory of Resonance and its Application to Organic Chemistry (New York, 1944), pp. 1-88.

XVlJl INTRODUCTION conditions its actual state is not given by either, nor by a mixture of the two in chemical (tautomeric) equilibrium, but is a hybrid between them, and has to some extent the properties of both. The conditions which must be satisfied for this resonance to be possible are three: (1) the positions of the atoms in the two structures must be nearly the same, (2) thp energy contents of the two (their relative stabilities) must not differ too greatly, the state of the hybrid being nearer to the more stable form, and (3) the number of unpaired (not of unshared) electrons must be the same in both; this last condition is almost always fulfilled. The resonance produces two important effects: (A) the energy content of the molecule is less, and its stability greater, than in either form, and hence the resonance must always occur when the conditions 1-3 are satisfied; (B) the linked atoms are rather closer together than in either separate form, owing to the greater strength of the link; this result is of great diagnostic importance in giving evidence of resonance.12

A typical example is that of carbon dioxide; the possible forms, with their distances and heats of formation from the atoms, are given below, and are compared with the observed values:

Distance . . .J IJ oat of formation from (

atoms . . .1

O=C=O 1-22+1-22

2-44 173 + 173

346

0 = C ~ > 0 1-37 + 1-10

2-47 81-5 + 256

337-5

0

INTRODUCTION X l X ionic character. If X1 and X2 are the values, the extra energy due to the partial ionic character is about 23(^1x2)2 k.cals./mole. With a difference of 1-7 units the link is about 50 per cent, ionic.

It should be noticed that these mixed ionie-covalent links have in some important respects the properties of covalent but not of ionic links. Two of the principal properties by which ionic links are recognized are (1) the independent motions of the ions (conductivity, low molecular weight in solution), and (2) their freedom of position in the molecule, leading to close-packing and the absence of isomers. Neither of these properties is to be found with the ionic-covalent bond; it is a condition of resonance that the atoms must occupy nearly the same places in both forms, so that the ions cannot separate; and as the positions are fixed in the covalent structure but adaptable in the electrovalent, the positions in the hybrid must be those of the covalent form. Thus for many purposes molecules with these mixed links must be regarded as covalent and not ionized.

STEREOCHEMISTRY1* The arrangements in space of the covalencies of poly-covalent atoms,

while they are subject to small variations seldom exceeding 5 or 10, tend to conform to one or other of quite a limited number of types. These are very simply related to the size of the valency group of electrons (in Lewis's sense, the shared electrons counting for both atoms) if the imaginary positions of the electronic pairs are taken to be the same whether they are occupied (shared) or not.

I. When the valency group is 4 we have with a covalency of 2 a linear structure (180) as in Cl--Hg-Cl.

II. When the valency group is 6, if they are all shared we have three covalencies and these are at 120 in a plane with the central atom, as has been shown in boron trifluoride. Where only 4 of the 6 are shared the valency angle is still about 120 (as in stannous chloride).

III . With a complete valency octet (apart from the transitional elements, which require special treatment) the arrangement is tetrahedral; if some pairs are unshared the positions of the rest are only slightly affected (the angles being usually reduced from 109*5 in the direction of 90) and are for a covalency of 3 (2, 6*) pyramidal, as in NH3, and for 2 (4,4) triangular, as in H2O.

With the transitional elements the size of the valency group is uncertain, m it may include any or all of the electrons above 8 of the previous (penultimate) group. It is found that when this maximum size (n) is not muoh more than 8, the structure of a 4-covalent atom is tetrahedral, and we may assume that there the valency group is a shared octet, all the

* Shared eleotrons are underlined. M

N. V. Sidgwiok and H, M, Powell, Proo. Boy, Soo* 1040,176, 158.

xx INTRODUCTION unshared electrons being in the previous group. When, however, n is not much less than 18, the 4-covalent structure is found to be planar; a reason for this is suggested later under V.

IV. When there are 10 valency electrons (the so-called inert pair (see p. 287) if present being counted in), the arrangement is that of a trigonal bipyramid, with two points at the poles of a sphere and the other three symmetrically disposed (at 120) on the equator (Fig. 1).

Trigonal Bipyramid FIG. 1

IF6 Structure FIG. 2

With 5-covalent decets (10) this has been established in every case examined, for example, with PF5, TaCl5, Fe(CO)5, and Sb(CH3)2(hal)3.

The 4-covalent decet (2, 8) has the structure derived from this with one of the five points empty. This has been established with K[IO2F2]16 and with TeCl416 (Fig. 2).

The 3-covalent decet occurs in the aryl iodide-chlorides Ar-ICl2, but their structure is unknown. The 2-covalent decet (4, 6) occurs in the trihalide ions, as in M[ICl2] and M[I8], which are known to be linear (2 polar points).

V. Duodecet. In its fully shared form AB6 this gives the octahedron established by Werner. On theoretical grounds17 the three structures below are all possible: of these II is really a form of I in which the B B distances

CXr---v.-

I . Trigonal Antiprism. I I . Octahedron. I I I . Trigonal Prism. FIG. 3

and the valency angles are all equal. Experimentally all AB6 molecules are found to have the octahedral structure I I except a few giant mole-cules (for instance, MoS2 III, and nickel arsenide I and III). The reason, no doubt, is that the octahedron gives for a fixed AB length the greatest distance from one B to another, and so is favoured, owing to the mutual repulsion of the B's, wherever the attraction of neighbouring atoms does not interfere, as it does in the giant molecules.

L. Helmholz and M. T. Bogrs, J.A.C.S. 1940,62, 1537. M

D, P. Steveneon and V. Sahomakor, ib. 1267. 17

Q, B, Kimball, J, Chmn, Phyt, IMO1S, 188,

I N T R O D U C T I O N xxi The 5-covalent duodecet (2, 10) must occur in IF5 (Fig. 2), but the

structure of this cannot be determined by electron diffraction because of the great difference (53 to 9) between the atomic numbers of iodine and fluorine.

The 4-covalent duodecet is found in the unusual type of M[ICl4], where the anion has been shown to have a planar square structure, i.e. an octa-hedron with two trans positions empty. I t may be supposed that the same 4, 8 electronic arrangement occurs in the planar 4-covalent derivatives of the later transitional elements, such as M2[Ni(CN)4]; the nickel here has the atomic composition 2, 8 (16, 8), which presumably should be written 2, 8, 12 (4, 8), giving the same type of duodecet as in M2[ICl4].

VI. 14*group. The very few examples of this rare condition that have been measured show two types of structure, one (in K3[ZrF7]) derived from an octahedron by adding a fluorine atom to the centre of one face,18 and the other (in K2[NbF7] and K2[TaF7]) from a trigonal prism by adding a fluorine atom at the centre of a prism face,19 the strain in both cases being eased by distortion.

VII. 16-group: covalency 8. Only one compound of this rare kind has been examined, the very stable octacyanide K4[Mo(CN)8]; the anion of this salt has been shown20 to have neither the cubic nor the antiprismatic (twisted cube) form, but that of a dodecahedron.

Multiple Links. With the octet the stereochemistry of these is well known;

the angles are Ai^ 125-15 and both B = A = B and B - A = B 180. The >B

positions of multiple links with larger valency groups are scarcely known theoretically, and not at all practically.

ATOMIC BADII By X-ray and electron diffraction, from the spectra, and in other ways,

the lengths (distances between the nuclei) of a large number of covalent links have been measured, and it has been found that they can be approxi-mately expressed as the sum of two values, one for each of the linked atoms, which are known as the atomic radii (a similar additivity is found to hold for ions, though the values are, of course, different). The observed lengths are, however, subject to small variations, of which the most important are those due to the multiplicities: in general the ratios of the links AB, A=B, A = B are roughly 1:0-9:0-8.21 Resonance shortens the distances by introducing an element of multiplicity into single links, and in addition by the shortening which resonance itself involves.22 There are

18 G. C. Hampson and L. Pauling, J.A.C.S. 1938, 60, 2702.

J. L. Hoard, ib. 1939, 61, 1252. 80

J. L. Hoard and H. H. Nordsieck, ib. 1939,61, 2863. 81

J. L. Kavanau, J . Ohem. Phya, 1944,12, 467. 88

H. A, Skinner, Trans. Far. Soo. 1945,41, 645.

xxii I N T R O D U C T I O N other modifying influences as well, which are not yet fully understood.23""5 Where the distances in a molecule have been measured, it is useful to be able to compare them with those derived from some standard series of values of the atomic radii. A list (Table V) is therefore given (p. xxix) of the most probable radii of the atoms in covalent and electrovalent links, and values taken from this are appended (as ' theory') to the measurements quoted in the text. This is merely to facilitate comparison, and the * theoretical5 values must not be supposed to have any special validity.

HEATS OF LINKAGE The heat of formation Hf of a molecule from its elements in their

(specified) standard statesfor example, graphite, hydrogen gas, solid iodinecan be ascertained thermochemically. If the heat of volatilization or sublimation of the substance is known, and further the heats of atomiza-tion of the component elements from their standard states, the algebraic sum of all these quantities gives Ha, the heat of formation of the gaseous compound from its atoms; this is expressed in k.cals. per g.-molecule. The deduction from this of the heats of formation of individual links is (except with diatomic molecules) to some extent a matter of convention. Thus with water we have the following values (all, of course, for the gas)

(1) 2 H + O = H0H+2x 1102 k.cals./mol. (2) H+OH = H - O - H -f 103-5 (3) H + O ==HO + 116-9

Here we have three different values for the HO link. The values (2) and (3) depend on the stability of the radical OH; we are not concerned with this but only with the relation between H2O molecule and its con-stituent atoms, i.e. with (1); our object is to get values such that their sum for all the links in a molecule gives its Ha as nearly as possible. Hence the value used for the heat of linkage is got by dividing the Ha of the normal molecule ABx by v, the number of links that it contains (here H0 = 110-2 k.cals.). With such values the additivity for molecules with several kinds of links is found to hold very nearly, if allowance is made for the resonance energy when this is to be expected; the value of the resonance energy is, in fact, usually obtained by subtracting from the Ha of the substance the sum of the normal ('theoretical') values for the links that it contains. There are other influences which cause small changes in Ha, seldom amounting to 5 k.cals.; these will be discussed as they occur (see, for example, pp. 501-5).

The Tables VI AD which follow give the most probable values for A (p. xxx) the heats of atomization of the elements from their standard states B (p. xxxi), the heats of formation Ha of single links C (p. xxxii), those of multiple links, and D (p. xxxii), the effects (where known) of change of valency on the heat of formation of the link.

33 V. Sohomaker and D. P. Stevenson, J>A.Q. 1941,63, 37.

a* W. Gordy, J, 0&m. Phy. 1847,11 St, 908. L. Pftuling, i/.4,CJ, 1947 69,141 (for motuU),

INTRODUCTION

TABLE I

International Atomic weights 1948

Aluminium Antimony Argon Arsenic Barium Beryllium Bismuth Boron Bromine Cadmium Caesium Calcium Carbon [Cassiopaeum = Cerium Chlorine Chromium Cobalt [Columbium = > Copper Dysprosium Erbium Europium Fluorine Gadolinium Gallium Germanium Gold Hafnium Helium Holmium * Hydrogen Indium Iodine Iridium Iron Krypton Lanthanum Lead Lithium Lutecium Magnesium Manganese Mroury

Symbol Al Sb A As Ba Be Bi B Br Cd Cs Ca C

Luteciui Ce Cl Cr Co

[iobium] Cu Dy Er Eu F Gd Ga Ge Au Hf Ho Ho H In I Ir Fe Kr La Pb I Li I Lu l Mg Mn Hg

\AL \No.

13 51 18 33 56 4

83 5

35 48 55 20

6 Tl]

58 17 24 27

29 6Q 68 63 9

64 31 32 79 72

2 67

1 49 53 77 26 36 57 82

3 71 12 25 80

At. Wt. 26-97

121-76 39-944 74-91

137-36 9-02

209-00 10-82 79-916

112-41 132-91 40-08 12-010

14013 35-457 52-01 58-94

63-54 162-46 167-2 152-0

19-00 156-9 69-72 72-60

197-2 178-6

4-003 164-94

1-0080 114-76 126-92 1931 55-85 83-7

138-92 207-21

6-940 174-99 24-32 54-93

200-61

Molybdenum Neodymium Neon Nickel Niobium Nitrogen Osmium Oxygen Palladium Phosphorus i Platinum Potassium Praseodymium Protoactinium Radium Radon Rhenium

! Rhodium Rubidium Ruthenium Samarium Scandium Selenium Silicon Silver Sodium Strontium Sulphur Tantalum Tellurium Terbium Thallium Thorium Thulium Tin Titanium Tungsten Uranium Vanadium Xenon Ytterbium Yttrium Zinc Zirconium

Symbol Mo Nd Ne Ni Nb N Os O Pd P Pt K Pr Pa Ra Bn Re Rh Rb Ru Sm Sc Se Si Ag Na Sr S Ta Te Tb Tl Th Tm Sn Ti W U V Xe Yb Y Zn Zr

At. No. 42 60 10 28 41

7 76 8

46 15 78 19 59 I 91 88 86 75 45 37 44 62 21 34 14 47 11 38 16 73 52 65 81 90 69 50 22 74 92 23 54 70 39 30 40

At. Wt.

95-95 14427 20-183 58-69 92-91 14-008

190-2 16-0000

106-7 30-98

195-23 39096

140-92 231 226-05 222 * 186-31 102-91 85-48

101-7 150-43 45-10 78-96 28-06

107-880 22-997 87-63 32-066

180-88 127-61 159-2 204-39 23212 169-4 118-70 47-90

183-92 23807

50-95 131-3 17304 88*92 65-38 91-22

INTRODUCTION XXV

Group SvJbgp. 37Rb 38Sr 39 Y 40Zr 41Nb 42Mo 43Tc 4 4 R u 4 5 R h 46Pd

4nJJT~ 48Cd 4 9 I n 50Sn 51Sb 52Te 5 3 1 54Xe

55Cs 56Ba 57La 58Ce 59Pr 60Nd 6111 62Sm 63 Eu 64Gd 65 Tb 66By 67Ho 68Er 69 Tm 70Yb 71Lu 72Hf

I 1 8

2 2 2 2 2 2 2 2 2

I 2 2 2 2 2 2 2 2 2

I 2 2 2 2

2 ! 2 I 2 2 2 2 2 2 2 2 2 2 2 2

I 2 *P

8 8 8 8 8 8 8 8 8

I 8 8 8 8 8 8 8 8 8

i 8 8 8 8 8 8 8 8 8 : 8 8 8 8 8 8 J 8 8 8

I 3 * i # < *

18 18 18 18 18 18 18 18 18 18 18 18 18 18 18 18 18 18

I 1 8 18 18 18 18 18 18 18 18 18 18 18 18 18 18 18 18 18

8

2 2 2

I 2 2 2 2 2 2

I 2 2 2 2 2 2 2 2 2

2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2

P 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6

6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6

4 d

1 2 4 5 6 7 8

10 10 10 10 10 10 10 10 10

10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10

/

1 2 3 4 5 6 7 8 9

io i i 12 13 14 14

8

1 2 2 2

I _ 1 2 2 2 2 2 2 2

2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2

5 P d

1 2 3 4 5 6

6 6 6 1 6 1 6 1 6 1 6 1 6 1 6 1 6 1 6 1 6 1 6 1 6 1 6 1 6 1 6 1 6 2

6 8 p

f l 2 2 2

2 ' 2 2 2 2 2 2 2 2 2 2 2 2 2

xxiv INTRODUCTION

TABLE I I

Atomic Structures I!

I H 2 H e

3 L i 4 B e 6 B 6 C 7 N 8 0 9 F

10Ne

H N a 12Mg 13Al 14Si 16 P 16 S 17Cl 18A

19 K 20Ca I 21Sc I 22Ti 23 V 24Cr 26Mn i 26Fe j 27Co ! 28Ni 29Cu 30Zn 31Ga ! 32Ge j 83As 34Se 36Br 36Kr

1 S

1 2

' 2

2 2 2 2 2

! 2 2

2 2 2 2 2 2 j 2 2

2 2 2 2 2 ! 2 2 2 2 2 2 2 2 ! 2 2 2 2 2

I 2 s p

1 2 2 1 2 2 2 3 2 4 2 6 2 6

2 6 2 6 2 6 2 6 2 6 2 6 2 6 2 6

2 6 2 6 2 6 2 6 2 6 2 6 2 6 2 6 2 6 2 6 2 6 2 6 2 6 2 6 2 6 j 2 6 2 6 2 6

3 \ s p d

1 2 2 1 2 2

2 3 2 4 2 6 2 6

2 6 2 6 2 6 1 2 6 2 2 6 3 2 6 5 2 6 5 2 6 6 2 6 7 2 6 8 2 6 10 2 6 10 2 6 10 2 6 10 I 2 6 10 2 6 10 2 6 10 2 6 10

I 4 I s p d f

I 2 2 2 2

j 1 2 2 2 2 1 J 2 I 2 1 2 2 2 3 2 4 2 5 2 6

I 5 I S j9 d /

xxvi INTRODUCTION TABLE II

Atomic Structures (cont*) Group Subgp, 73Ta 74 W 76Re 76Os 77I r 7 8 P t 79 Au 80Hg 81Tl 82Pb 83Bi 84Po 86 At 86Em

87Fr 88Ra 89 Ao 90 Th 91 Pa 92 U

93Np 94 Pu 96 Am 96Cm

i 1 S

! 2 2 2 2 2 2

2 2 2 2

2 2 2 2

2 2 2 2 2 2

2 2 2 2 i

2

8 8 8 8 8 8

8 8 8 8 8 8 8 8

8 cS 8 8 8 8

8 8 8 8 i

3 \s,p9d

18 18 18 18 18 18 18 18 18 18

18 18 18 18

18 18 18 18 18 18

18 18 18 18

4 \s>P>d,f

32 32 32

! 32 32 32

j 32 ! 32

32 32 32 32 32 32

32 32 32 32 32 32

32 32 32 32 !

5 \ s p d f

2 6 3 ' 2 6 4

2 6 5 2 6 6 2 6 7 2 6 8

! 2 6 10 2 6 10 2 6 10

( 2 6 10 2 6 10 2 6 10 2 6 10 2 6 10

2 6 10 2 6 10 2 6 10 2 6 10 2 6 10 2 6 10 3

2 6 10 4 2 6 10 6 2 6 10 6 2 6 10 7

6 \ s p d f

2 2

! 2 ! 2

2 2 1 2 2 1 2 2 2 3 2 4 2 5 2 6

2 6 2 6 2 6 1 2 6 3 2 6 4 2 6 2

2 6 2 2 6 2 2 6 2 2 6 2 j

7 S

1 2 2

The structures assigned to the last few elements are speculative; the distribution of the electrons between the 5/ and the 6d orbits is uncertain; gee W. F. Meggers, Science, 1947,105,514, In these elements the quantum groups of the electrons are less important than the energies of the orbits, which are often very similar.

I N T R O D U C T I O N

TABLE III

Periodic Classification (Thomsen-Bohr) H i

/ \

He 2

/ \ / U Be B C N O F IsTe 3 4 5 6 7 8 9 10

Na Mg Al Si F S Cl A 11 12 1,3 14 1$ 16 17 18

K Ca 19 20

Bb Sr 37 38

Sc Ti V Cr Mn Fe Co Ni 21 22 23 24 25 26 27 28

Cu Zn Oa Ge As Se Br Kr 29 30 31 32 33 34 35 36

Y Zr Nb Mo Tc Ru Rh Pdj 39 40 41 42

Ag Cd In Bn Sb Te I Xe " '~ 49 SO 51 52 53 54

CsBa La 55 56 57

C e F r - Y b 58 59 70

Fr Ra 87 88

LuHfTaWReOsIrFt I 71 72 73 74 75 76 77 78 Au Hg Tl Fb Bi Fo At Em 79 80 81 82 83 84 85 86

Ac Th Pa TJ 89 90 91 92

Np Pu Am Cm 93 94 95 96

INTRODUCTION

TABLE V

Atomic and Ionic Radii in AXJ. (Atomic for single links unless otherwise stated)

Atomic H He Ne A Kr Xe

Li Na K Rb Cs

Cu Ag Au

Be Mg Ca Sr Ba Ra

Zn Cd Hg

B - -B = B=E= Al Sc Y La Ce Pr Nd

Sm Eu [Eu" Gd Tb By Ho Er Tm

030 1-79 1-60 1-91 1-97 218

1-33 1-57 203 216 235

135 1-53 1-50

0-90 1*36 1-74 1-92 1*98

1-31 1*48 1-48

0-88 0-76 0-68 1-26 1-44 1-62 1*69 1-65 1-65 1-64

1-64 1-63 1-85] 1-61 1-59 1'59 1*58 1-57 !56

Ionic

"Ii1

'

Li+ Na+ K+ Rb+ Cs+

Cu+ ca Ag+

Be++ Mg++ Ca++ Sr++ Ba++ Ra++

Zn+^ Cd++ HjB*++

Al+++ Sc+++ Y-HH-

La+++ Ce+++ pr+++ Nd+++

Sm+++ Eu+++

Gd+++ Tb+++ Dy+++ Ho+++ Er+++ Tm+++

1-3

0-78 0-98 j 1-33 149 1-66

1-0 j 1-13 I

034 0*78 1-06 1-27 1*43 1-52

0*83 103 112

0*83 0*78 0*93 1*22 1*18 1*16 1*15

113 1*13

111 x 109

107 105 1-04 1*04

Atomic Yb [Yb Lu

Ga In Tl

C -C== P = Si^-S i = Si=E= Ge G e = S n -S n = P b -

Ti Zr Hf Th

N N = N=s P P = P = As A s = S b -B i -

I V Nb Ta

O I O = ! O = j S

S== S s S e -

1*56 1*70] 1*56

1*26 1-44 1*47

0*771 0*665 0*602 117 107 1*00 1*22 112 1*40 1-30 1-46

1*36 1*48 1*48 1*65

0*70 0*60 0*55 110 1*00 0*93 1*21 111 1*41 1*46

1*23 1*34 1*34

0*66 0*55 0*50 1*04 0*94 0*87 1-17

Ionic

Yb+++

Lu+++

Ga++t In+++ Tl+++ Tl+

Sn4+ Pb4+ Pb2+ Ti4+ Zr4+

Th4+

N~

pa-

As3~

Sb8-

o

S "

S e "

1*00

0*99

0*62 0*92 1*05 1*49

0*74 0*84 1*32 0*64 0*87

110

1*7

2 1

2*2

2*4

1*32

1*74

!91

TABUS IV *

Periodic (Jlmdficctiion (Mendeleejf) Otoup

Pteriodl. 2. 3. 4.

5.

6.

7.

I I A. B. I

H l UB N a i l K 19

Cu 29 Rb 37

Ag 47 Cs 55

Au 79 Fr 87

11 I

A. B. !

Be 4 Mg 12 Ca 20

Zn 30 Sr 38

Cd 48 Ba 56

I Hg 80 Ra 88

III A. B.

B 5 Al 13 Sc 21

Ga 31 Y 39

In 49 La 57 Ce 58 Pr 59 Nd 60 1161 Sm* 62 Eu 63 Gd 64 Tb 65 Dy 66 Ho 67 Er 68 Tm 69 Yb 70 Lu 71

Tl 81 Ae 89

IV A. B.

C 6 Si 14 Ti 22

Ge 32 Zr 40

Sn 50 (Ce 58)

Hf 72 Pb 82

Th 90

A. B.

N 7 P 15 V 23

As 33 Nb 41

Sb 51

Ta 73 Bi 83

Pa 91

VI A. B.

0 8 S 16 Cr 24

Se 34 Mo 42

Te 52

W 74 Po 84

U 92 Np 93 Pu 94 Am 95 Cm 96

VII A. B.

F 9 Cl 17 Mn 25

Br 35 Tc 43

1 5 3

Re 75 At 85

VIII

Fe 26 Co 27 Ni 28

Ru 44 Rh 45 Pd 46

Os 76 Ir 77 Pt 78

0

He 2 Ne 10 A 18

Kr 36

Xe 54

Em 86

xxx INTRODUCTION

TABLE V (cont.) Atomic

S e = 1-07 T e - 1-37 T e = 1-27 P o - 1-70 Cr 1*25 Mo 1-36 W 1-37 U 1-49

F 0*64 F = 0-54 C l - 0-99 C l = 0-89 Br 1-14 B r = 1*04 I 1-33

Ionic

T e Te4+

Cr3+ Mo4+ W4+ U4+

F -

Cl-

Br-

I~

211 0-89

0-65 0*68 0-68 1-05

1-33

1-81

1-96

2-20

I Atomic

I I =

Mn Re

Fe" Fe'" Co" Co"' Ni"

Ru" Rh'" Pd* Os* Irw

1*23

1*25 1*37

1*23 1-22 1*32 1*22 1-39

1*33 1*32 1-31 1*33 1*32

Ionic

Fe++ Fe+++ Co++

Ni++

Ru4+ Rh+++

Os4+ Ir4+

0-83 0*67 0-82

0*78

0*65 0*69

0*67 0*66

TABLE VI A

Heats of Atomization from Standard States; h.cals per g.-atom H 51*7 Li 39*0 Na 25*9 K 19-8 Rb 18*9

Be

Zn Cd Hg O S Se

75

27*4

14*6 591 66*3 61*0

B Al

Tl

115 55

40

C 170*4 Si 85*0 Ge 85*0 Sn 78*0 Pb 47*5 Ti 100 Zr 110

F 16*7

INTRODUCTION

TABLE VI B

Heats of Formation of Links from Atoms (Ha) H

H C N O S F Cl Br I

Ag Be Zn Hg

B Al Tl Si Ge Sn Pb" Ti Zr

P As V Se

Te Fe As

103-4 98-8 92-9

110-2 87-5

132-4 102-7 87-3 71-4

751

77-3 56-7

730

62-3

G 98-8 81-6 69-3 81-5 66-0

103-4 78-0 65-5 57-0

34-5 ca* 15

69-1

C 70-7 = C 165-6

N 92-9 69-3 38-4

62*9 47-6

O 110-2 81-5

34-9

43-5 49-3

60-8

91-2

89-3

65-1 110-4

(OsO4)

8 87-5 66-0

63-8 71-4 66-1 57-2

60-9

F 132-4 103-4 62-9 43-5 71-4

33-3 71-3

140-6

127-9

89-9

67-1

78-3

Cl 102-7 78-0 47-6 49-3 66-1 71-3 57-8 52-7 51-0 70-4

107-9

52-9 96-7 91-6 85-3 85-8

104-1 78-0 77-4 97-0

116-9 77-1 69-7 88-7 66-8

Br

87-3 65-5

57-2

52-7 46-1 42-9 67-7 88-4

450 76*9

76-4 69-3

66-2 53-2

63-5 57-4

I

71-4 57-0

510 42-9 36-2 60-1 68-6

35-4

66-3 51*1

44-4

49-5 42-5

SiSi 42-5 Ge-Ge 42-5 P - P 149-2 AsAs 117-2 Se-Se 57-6

I N T R O D U C T I O N

TABLE VI C

Multiple Links

C - C C = C C = C N - N N = N N=E=N

oo 0 = 0

Abs. 81-6

1461 1921 38-4 97-6

2250 34-9 960

Bel. I 1-00 1-79 2-35 100 2-54 5-86 100 2-75

S - S S = S S e - S e Se=Se PP& P = P AsAs6 AsE=As

Abs. 63-8

103-4 57-6 92-5 47-5

1170 34-3 91-3

Bel. 1-00 1-62 1-00 1*63 100 2-46 1-00 2-69

C - N i C = N

C E = N C - O C = O CE=O C - S C = S

Abs. 69-3

1350 2120

81-5 173-0 256-0

66-0 1260

Bel. 1-00 1-95 3-06 100 2-12 314 1-00 1-91

a For the 1JD state.

* See H. A. Skinner, Trans. Far. Soc. 1945, 41 , 645; F . S. Dainton, ibid. 1947, 43 , 244.

TABLE VI D

Effect of Valency P - C l in PCl8 77-1

in PCl6 62-0 Ratio 0-80

ICl in ICl 51-0 in ICl8 42-0

0-82

A l - C l in AlCl3 91-6 in Al2Cl6 73-7

0*80

S b - C l in SbCl3 68-0 in SbCl5 58*6

Ratio 0-82

GROUP 0 INERT GASES

THE inert gases occupy a peculiar position in Chemistry. They are practically devoid of chemical properties, and yet for that very reason they have provided the key to the whole problem of valency and the interpretation of the Periodic Classification.

Further, their physical behaviour, largely owing to this inertness, is of great interest. Helium in particular shows physical properties to which there is no parallel elsewhere, and it has opened up quite a new field of physical investigation, that of extremely low temperaturethe 'uncharted infinity' that lies between 1 K, and the absolute zero.

The more important physical properties of the inert gases are given in the following table.

TABLE

B.-pt. K. C.

M.-pt. K. 0C.

(hit.T. K. C.

Vol. in c.c. in L m.3 airf

JIt. of evapn. ,, fusion

(k.eals./g.at.)c M.-pt./Crit.T.6 Trouton Afc. Diam. in

Crystal (A)

He

4-216a -268-94

. .

5-236 -267-93

6-24* 0-0196 0-00436 (3 K.)

4-64 3-67*

Ne

27-17 -~ 245-99

24-506 -248-7

44.70& -228-5

1821 0-42 0-801

0-548 19-8 3-20

:

A

87-92 -185-24

83-86 -189-4

150-69 -122-47

9,300 1-59 0-2808

0-556 18-1 3-82

Kr

120-9 -63-8

1

2138* 0-3907*

0-554 17-7

3-94

Xe

165-1 -108-1

161-3c -111-9

289-90& + 16-7

0 1 3-020 0-5485c

0-557 18-3 4-36

Em

208 - 6 5

160 -113

377-6 + 104-4

6 XlO"14

a = \ b = 2, c = 3, d = \ e = B, / = 6, g = 7, h = 8, i =

HELIUM This element was discovered spectroscopically in the chromosphere of the

tun in 1868 by Loekyer and Frankland, who therefore called it helium. In ISB5, after the discovery of argon, Ramsay found that the inactive gas which Hlllebrand had obtained in 1890 from cleveite and other uranium minerals by heating or solution gave the same spectrum, and so must be helium.

1 G. Schmidt and W. H. Keesom, Physica, 1937, 4 , 963.

1 K. Clusius and K. Weigand, Z.physikal. Chem, 1939, B 42, 111.

8 K. Clusius and L. Riccoboni, ib. 1938, B 38, 81.

* JB. Justi, Phys. Z. 1935, 36, 571. 1 K. Clusius, A. Kruis, and F . Konnertz, Ann, Phys, 1938, [v] 33 , 642.

fl F . A. Paneth, Nature, 1937, 139, 181.

f H. GUiokauf, Proc. Boy. Soc. 1946, 185, 98.

1 W. H. Keesom and K. W. Taoonis, Proc, K. Ahad. Amst. 1938, 4 1 , 95.

K. Glusiui, Z, physikal Ch&m, 1936, B 31, 459.

2 Group O. Helium . Helium is produced in the decay of radioactive elements (1 kg. of uranium in its conversion into 865 g. of lead forms 135 g. or 756 litres of helium), and also by the bombardment of lithium, beryllium, and other (mainly light) elements with projectiles or rays of high energy, such .as cosmic rays, X-rays, and high-speed protons and deuterons. These bombardments must take place in nature, and give a continuous supply of helium.

On the other hand, owing to its lightness it will tend to escape from the earth's atmosphere. A molecule can do this10 if it has a velocity of 11*2 km./sec. (on the moon, of 24 km./sec). Thus the nearer the mean velocity is to this escape velocity, the more rapidly the gas will escape. The mean velocities at 0 C. are:

Km./see. Neutron

2'60 H2 1-84

He

1-30 Ne

0-58 A

0-41 Kr

0-28 O2

0-46 CO2 0-39

It can be shown that if the mean velocity of any gas is one-third of the escape velocity (i.e. on the earth 3*7 km./sec), half of the gas will escape in a few weeks. It is therefore evident that the helium will escape fairly rapidly, though not so rapidly as the hydrogen or the neutrons, while the other inert gases will be retained. As Paneth has pointed out, helium, in spite of its name, is the only (non-radioactive) element on the earth that did not originate in the sun. All that we now have must have been born on the earth since its separation.

The technical source of helium is certain natural gases, especially oil gas in America and Canada, which often contains from 0-1 to 2 per cent. From this source the U.S. Government had obtained by 194O11 nearly 150 million cubic feet, mostly at Amarillo, Texas: the price then was less than 1-5 cents per cubic foot. This has been used for airships, since it is not inflammable. Its lifting power as compared with hydrogen is

oW-*?*=*). \ 28-8-2/

A large airship will contain 2-3 million cubic feet. Isotopes

Four isotopes of helium are known: 4He, which forms practically all the natural element, and 3He, 6He, and 0He, which can be made artificially, and of the first of which there may be a trace in the natural gas.

3He was found by Bleakney12 to be formed when deuterium is bom-barded by deuterons; two reactions occur13:

2D + 2D ~> 3He + 1Ti and -> 3H + 1H 10

H. N. Bussell, Nature, 1935, 135, 223. 11

J. Ind. Eng. Chem. (News Ed.), 1940, 830. 12

W, Bleakney, G, P. Hamwell, W, W, Losier, P. T. Smith, and H. D. Smyth, Phys. Mev. 1084, 46, 81.

18 M. L. E, Oliphwt, P, Harttok, and Lord Butherford, Proo. Roy, Soc. 1934,

144 0Oi,

Helium 3 the two resulting nuclei (composed respectively of two protons and one neutron, or of one proton and two neutrons) both weigh 3-017213*15; both are unstable, though they are relatively long lived.

This 3He has recently been found to be present in minute quantities in natural helium; Alvarez and Cornog,16'17 using a 60-inch cyclotron as a mass spectrograph, have got evidence that it occurs to an extent of 10~7 in atmospheric helium and 10~8 in helium from gas wells.

The isotope 6He is far more unstable. It is formed in several nuclear reactions; it is produced18 when 'heavy* paraffin is bombarded with a-particles:

2D + 4 ~> 6He + 1H. It can also be made19'20 along with ordinary helium by bombarding lithium with swift deuterons:

7Li + 2D ~> 4He + 5He. Its mass is 5-010621; it rapidly decays thus:

6He -> 4He + %. 0He has a half-life of 0-85 sec.22 The following are the masses of the three isotopes:

3He

30172 3x10057

4He 4-00380

4x1*00097

5He

5-0106* 5x10021

a = *4, 6 = 21

From these figures it is evident that if four hydrogen atoms (4 X 1-00813 -~ 4*03252) could be made to form a 4He nucleus, they would lose 0*02866 of a unit of mass, and hence emit 0*02866 X 940 = 26-94 m.e.v. of energy; this is 622 million k.cals. for 4 g. of hydrogen, or 156 millions per gramme. Bethe has shown25 that this could be brought about in a star by the follow-ing series of reactions:

1IC + IK -> "KT; \3N ~-> 1IC + ^ ; 1JC + JH ~> \4N; 1JN + JH ~> 1IO; 1JO -> 1JN + e+; X?N + JH - 1JC + 4He.

u N. R. Sen and U. B* Burman, Astrophys. J. 1944, 100, 347.

16 T. W. Bonner, Phys. Rev. 1938, 53 , 711.

16 L. W. Alvarez and B . Cornog, ib. 1939, 56, 379.

" Ib. 613. 1H

F. Joliot and I. Zlotowski, CR. 1938, 206, 17, 1250. , e

J . H. Williams, W. G. Shepherd, and B . O. Haxby, Phys. Rev. 1937, 5 1 , 888. H. Siaub and W. E. Stephens, ib. 1939, 55, 845. ai

P. Joliot and I. Zlotowski, J . Phys. Radium, 1938, [vii] 9, 403. 98

H. S. Sommers and B . Sherr, Phys. Rev. 1946, ii. 69, 21.

B J . H. Jeans, Nature, 1943, 151, 7.

M K. T. Bainbridge, Phys. Rev. 1938, 53 , 922.

* H. A, Bathe, ib. 1939, 55 , 434. 86

H. N. Standi , J . Franklin ImU 1939, 228, 143.

4 Group O. Helium the carbon acting as a catalyst; and that these could occur in the interior of stars quickly enough to give the stellar temperatures required by astrophysicists (e.g. sun 18-21, Sirius 22 million 0C.).26

The rate of these nuclear reactions under bombardment is very sensitive to temperature, and has a sharp maximum at a critical temperature which is higher the heavier the nucleus attacked. It can be shown23 that as a star condenses under its own gravitation after separation from its galaxy, its temperature rises. When it reaches 3-7 million degrees, the Li, Be, and B nuclei are attacked by the protons, which is no doubt why these elements are abnormally rare. After they have mostly disappeared, a further rise of temperature occurs, and at about 20 million degrees, which actually is the approximate central temperature of the largest class of stars ('main-sequence' stars), the carbon nuclei are attacked, and the chain of reactions given above takes place, in which the carbon (unlike the Li, Be, and B) is not used up, but merely catalyses the conversion of the hydrogen into helium+energy. The amount of hydrogen being large, this process will last for some time; the sun is now emitting 250 million tons of energy a minute, and if it had only per cent, of hydrogen (it certainly has more) the energy of its conversion would last at this rate of loss for 2,000 million years.

When the hydrogen has at last disappeared, the star will begin to con-tract and warm up again, unless or until other nuclear reactions become possible in sufficient amount to balance the emission of energy. This is the condition of the very small and hot stars, whose central temperatures are found to be much above 20 million degrees. (See further Sen and Burman.14)

The chief scientific use of helium, the most volatile of liquids, is for the production of low temperatures by its evaporation under reduced pres-sures. The process was developed by Kamerlingh Onnes and Keesom at -Leyden, until finally Keesom succeeded, by the use of a large battery of pumps, in reaching 0*71 K. by keeping the pressure down to 0*00036 mm. of mercury. This is about as far as the process can be usefully carried. But much lower temperatures can be obtained by a method suggested by Debye and by Giauque, and carried out by Giauque, de Haas, and Simon, In a paramagnetic substance the atoms, or some of them, act as little magnets, which are normally oriented at random. When it is magnetized, these are all oriented parallel to one another, and this increase of order (or fall of entropy) must, if the substance is thermally isolated, be accom-panied by a diminution of order in the atomic motions, that is, by a rise of temperature. Conversely when it is demagnetized by removing the magnetic field, heat is absorbed. The substance used must remain para-magnetic down to the lowest temperatures reached: Giauque used gado-linium sulphate, but Simon and de Haas have found that one can use much cheaper materials such as ammonium iron alum and potassium chromium alum, The salt is enolosed in a metallic vessel surrounded by a vacuum, the magnetio field Ii turned on, and liquid helium added, The helium gas

Liquid Helium 5 is then pumped off until the vessel has fallen to 1 or 2 K., and the magnetic field removed. In this way Simon, using a field of 14,000 Gauss, has reached 0-03 K., while de Haas, with a much stronger magnet, claims to have reached 0-005 Abs. To realize the result we must remember that it is the ratio of the temperatures that matters, and not their difference. The change from 0-005 K. to 4-22 K., the boiling-point of helium, is an increase in the ratio 844 to 1; another step upwards of the same size would bring us to 4-22 X 844, or 3,562 K., nearly the boiling-point of carbon.

The success of this process is helped by the facts that the specific heat of the metallic container is almost negligible at these low temperatures (even at 12 K. 1 c.c. of helium under 100 atmospheres has as large a capacity for heat as 1 kg. of copper), and that owing to the low pressure of the gas (3 X 10~16 mm. at 0-2 K.) the vacuum jacket is an almost perfect insulator.

These temperatures are measured down to about 1 K. by a helium gas thermometer under low pressure, and below this by means of the mag-netic susceptibility (to weak fields) of the paramagnetic material, which is approximately proportional to the reciprocal of the absolute tem-perature.

Liquid and Solid Helium* In the liquid and solid states helium has properties unlike those of any

other substance. Liquid helium occurs in two forms, He I and He II, with a sharp transi-

tion point (the A-point) at 2-186 K. under 3-83 cm. mercury27; this falls as the pressure rises, and the triple point for solidHe IHe II is 1-774 K. under 28-91 atmospheres.28 He I (above this temperature) is a normal liquid; He II, below it, is unlike any other known substance. He II ex-pands on cooling; it has 10 times the specific heat of He I, but this rapidly falls; its conductivity for heat is enormous, being found (by the usual methods of measurement) to be about 3 million times that of He I, and nbout 200 times that of copper at the ordinary temperature. Again, while the viscosity of He I is normal, and that of He II as measured by a rotating disk,29 though it drops rapidly with falling temperature, retains a finite Value, the viscosity of helium II measured in a capillary of fine bore or by flow through a narrow slit appears to be zero30 or at least less than 10 1l poises.

Htill more remarkably, neither the heat conduction nor the viscosity bay what are otherwise universal rules: the heat transport is not

* I am greatly indebted to Dr. K. Mendelssohn for his help in dealing with this ttbjoot.

M G. Schmidt and W. H. Keesom, Physica, 1937, 4 , 971. w J . J . van Laar, Proc. K. Akad. Amst. 1936, 39, 612, 822. W. H. Keesom and G. E. Maowood, Physica, 1938, 5, 737. P. L. Kapitssa, Nature, 1988, 141, 74: J . F , Allen and A. D. Misener, Proc. Boy.

fee. 1089, 172, 467.

6 Group O. Helium proportional to (i.e. the conductivity is dependent on) the temperature gradient, and the flow through a fine capillary is independent of the head, though it changes rapidly with temperature. I t has further been shown31 that when the liquid flows from a higher to a lower level through a capillary it is cooled.

These very remarkable facts have been explained up to a point by an observation of Allen and Jones,32 who found that when one of two con-tainers filled with liquid helium I I and connected by a capillary is warmed (always below the A-point), helium will flow through the capillary towards the higher temperature. By using a tube filled with emery (i.e. with many fine capillaries) heated in the upper part by radiation from a lamp outside, a jet of liquid helium 3 or 4 cm. high can be got. The effect diminishes as the tube gets wider, and vanishes at a diameter of about 1 mm. This anomalous flow seems to give rise to convection currents which are free of friction in one direction. For instance, if a small flask with a heating wire near the bottom is immersed horizontally in HeI I and the wire heated, a stream of helium flows out through the neck and will move a vane, though no helium appears to move in.33 For some reason helium in this state can form a layer on any solid surface about 5 X 10~6 cm. thick (500 A.U., or, say, 140 atoms) which can move on the surface without any measurable friction. The occurrence of this flow is easily demonstrated.34*35 If the bottom of an empty vessel is dipped into He II it fills up to the same level; if it is then raised the level falls to that of the liquid outside; if the vessel when full is lifted out of the helium, the liquid drips from the bottom until it is empty. The rate of flow in c.c./sec. for every cm. of contact with the walls is 2-5 X 10~5 at 2 K., and about 7-5 x 10~5 at 1 K.: it is independent of the difference of height (the rate of movement of the film at 1 K., if it is 5 x 10~6 cm. thick, must be about 20 cm./sec).

The Debye X-ray pictures show that helium is not a crystalline liquid * at any temperature from 1*12 to 4-22 K.,36 and there is no change at the A-point in refractive power,37 molecular volume, or surface tension.38 At very low temperatures many of these peculiarities disappear. For a general account of these phenomena see Darrow.39 The physical structure underlying and explaining these peculiarities is not clear. Daunt and Mendelssohn40 point' out the analogy between the properties of He I I

31 J . G. Daunt and K. Mendelssohn, Nature, 1939, 143, 719; P. L. Kapitza, Phys.

Rev. 1941, 60, 354. 82

J . F . Allen and H. Jones, Nature, 1938, 141, 243. 88

P . L. Kapitza, J. Phys. U.S.S.R. 1940, 4, 181; 1941, 5, 59. 34

J . G. Daunt and K. Mendelssohn, Nature, 1938, 141, 911; 142, 475. 35

Id., Proc. Roy. Soc. 1939, 170, 423, 439. 80

W. H. Keesom and K. W. Taconis, Physica, 1937, 4, 28, 256 (the latter a small correction).

87 E. F . Burton, Nature, 1637, 140, 1015.

88 J. F, Allen and A. D, Misoner, Proc, Oamb* Philoa, Soc, 1083, 34, 209,

8 K. K. Darrow, Mm Mod. Phyi, 1040,13, Ul'.

40 J. G, Daunt and K. Mendelssohn, Nature, 194I1 IBOi S04.

Solid Helium 7 (frictional transport up to a critical rate of flow) and the phenomenon of super-conductivity; they say that He I I must contain atoms of low or zero thermal energy, separated by their velocities from the rest. A theory of liquid helium I I based on the condensation of a Bose-Einstein gas has been proposed by F. London42 and developed by Tisza.41 The theory of Landau43 follows similar lines but is not based on a special model. Accord-ing to this there are two forms of liquid helium, one (N) a normal liquid, and the other (S) with zero entropy and no viscosity. Down to 2-19 K. the liquid is all N; below this S appears in increasing quantity until at 0 K. it is all S. Thus when the flask (p. 6) is warmed there is an outward flow of N" and an inward flow of S, but the latter does not affect the vane since it has no viscosity; when this S reaches the heater it is changed into N". The heat conductivity is due to the fact that heat must be expended to transform the zero-entropy from S into N; the cooling on passage through a capillary is due to a kind of filter effect, the S passing through more easily than the N, and being then converted into N with absorption of heat.

The theories of Tisza and of Landau both predict that in He II tempera-ture differences should be propagated in the form of wave motion with a characteristic temperature-dependent velocity. This so-called 'second sound' has indeed been observed.44

Solid Helium Helium is the only liquid which cannot be frozen by lowering the

tomperature; at the ordinary pressure it must remain liquid down to the absolute zero. On the other hand, it is readily frozen by increasing the pressure; its freezing-point under various pressures has been found45'46 to be:

Temp. 0K. 1 2 3 10 20 42 Press., atm. 25 35 75-5 590 1,740 5,450

This is due to the czero-point energy'. We now know that substances still retain, even at the absolute zero, half a quantum of energy in every degree of oscillational freedom. This can be deduced from Heisenberg's Uncer-tainty Principle; it is really no more surprising than that a hydrogen atom at the absolute zero should still have its electron in the first quantum ltate and not in the nucleus. The result is that the atoms in liquid helium continue at the lowest temperatures to oscillate, and the energy of this mv\llation is greater than the heat of fusion; their motions are too lively far them to form the crystal. It is only when these motions are restrained by external pressure that the liquid can solidify. The effect is still per-

L. Tisza, Nature, 1938, 141, 913. F. London, Phys. Rev. 1938, 54, 947, L. Landau, ib. 1041, 60, 356. w V. Poehkov, J. Phys. U.S.S.B. 1946, 10, 389. W W, H. Ke@som Nature, 1926, 118, 81; C,R. 1926, 183, 26, 189. 41

F. Simon, M, Ruhomann, and W. A. M. Edwards, Z. phyeikah Chem, 1929,

8 Oroup O. Inert Oases ceptible even in the crystal, as the interatomic distances show (He 3-57; Ne 3-20; A3-82A.U.).

NEON, ARGON, KRYPTON, XENON, EMANATION Of the later members of this group there is little to say, apart from a

few special points, except on the obscure question of the formation of chemical compounds. Their more important physical properties have already been given in the table on p. 1.

Neon is used, generally mixed with helium, in the familiar neon lamps. The efficiency (light/energy) of these is about four times as great as that of the best metallic filaments, and nearly 25 per cent, of the theoretical maximum, at which all the energy appears as light.

Argon*7 is used for gas-filled incandescent lamps, mainly to prevent the evaporation and sputtering of the filament, and is usually mixed with about 15 per cent, of nitrogen to stop the formation of an arc.

Emanation, The importance of the three isotopic emanationsnow known by the rather unpleasant names of radon, actinon, and thoron in the discovery and development of the disintegration theory of radio-activity, and in the explanation of the periodic classification, is obvious. Their atomic weights and half-lives are:

Atomic weight . Half-life . ReL half-life .

Actinon 219

3*9 sec. 1

Thoron 220

54-5 sec. 14

Radon 222

3-8 days 84,200

Compounds of the Inert Gases The compounds or supposed compounds of the inert gases are of three

kinds: (1) molecules and molecular ions of the inert gases themselves; (2) supposed compounds of uncertain composition with other elements, mainly metals; (3) solid phases of definite stoichiometric formulae formed with certain other molecules such as water.

I. Ions of the He^ type are undoubted diatomic molecules; He2J" is formed by the attachment of a metastable to a normal atom; the heat of dissociation is 108-4 k.cals./mole.48 As long as they remain charged gaseous ions they are reasonably stable,49*50 though they cannot act as the cations of salts; the atoms are 1-09 A apart (which should be compared with the He- -He distance in solid helium of 3-57 A).

47 For entropy, etc., of argon (triple pt. 83-78 K. at 517 mm.) see K. Clusius and

A. Frank, Z. Elch. 1943, 49, 308. 48

F. L. Arnot and M. B. M'Ewen, Proc. Boy. Soo. 1939, 171 106. 49

L. Pauling, Chemical Bond, 1989, p. 240. 80

E, Majorana, Nuovo Om1 1981, 8, 22; L, Pauling, J*. 0/usm. Phy. 1988, 1, 56; . Weinbttiirn, ib, 1935, 3, 8471 L. Pauling and B. B. Wllion, Introd, to Quantum M$ohm4Q9t 901.

Compounds of Inert Gases 9 II. Various metals have been stated to form compounds with inert gases

(practically only with helium), usually on sparking; when solid, the pro-ducts have been found to give off their helium on heating or dissolving. Manley51 describes a volatile compound with mercury, detected by the spectrum. Boomer52 obtained with tungsten a solid of the composition WHe2, decomposed by heat. Damianovich53-5 found that on sparking helium with platinum a brown deposit was formed containing up to 34 c.c. of helium to 1 g. metal (Pt3#3He); he obtained similar substances with bismuth and uranium.56-7 None of these substances have been shown to be definite compounds, and it is at least possible that the solids merely contain adsorbed helium on the dispersed metal; indeed, Damianovich says58 that his platinum product gave X-ray diagrams resembling those given by colloidal platinum.

III. Substances of the third class are formed by the inert gases other than helium, and the more easily the heavier the gas, with water, D2O, phenol, and boron trifluoride. There is no doubt about the existence, nor usually about the composition and formulae of these substances; but there is no evidence of their occurring in anything but solid state, and so they are presumably van der Waals crystal aggregates, like the solid hydrates of methane and methyl bromide.

The hydrates were discovered by de JForcrand,59 who gives these com-positions, melting-points, and dissociation tensions at 0 C.:

Melting-point . Diss. tens, a t 0 C. .

A, a;HaO 80C.

98 atm.

Kr, 5H2O 130C.

15-5 atm.

Xe, a;HaO 240C.

1-3 atm.

Tammann and Krige60 confirm these data for the solid Kr, 5H2O. Deuterates of similar types, with the compositions Kr, 6D2O and Xe, 6D2O were made by Godchot et al.61

Nikitin62"4 showed that the solid SO2, 6H2O will absorb the inert gases other than helium if it is shaken with them, a definite ratio being estab-lished between the concentration of the inert gas in the solid and the

61 J . J . Manley, Phil. Mag. 1927, vii. 4 , 699.

62 E. H. Boomer, Proc. Roy. Soc. 1925, 109, 198.

63 H. Damianovich, CR. 1929, 188, 790.

54 H. Damianovich and J . J . Trillat, ib. 991.

56 H. Damianovich, Anal. Quim. Argentina, 1929, 17, 95.

56 Id., Anal. Inst, invest, dent, techn. 1934, 3/4, 20.

67 For a summary of the evidence for the formation by He, A, and Xe of com-

pounds with metals see H. Damianovich, Proc. 8th Amer. Sd. Congress, 1942, 7, 137. 68

H. Damianovich, Anal. Inst, invest, dent, techn. 1931, 2 , 15, 24. 60

R. de Forcrand, CR. 1923, 176, 355; 1925, 181, 15. 10

G. Tammann and G. J . K. Krige, Z. anorg. Ghent. 1925, 146, 179. t l M. Godchot, G. Cauquil, and R. Calas, CR. 1936, 202, 759.

8 B. A. Nikitin, Nature, 1937, 140, 643.

Id., J. Qm. Chem, Buss. 1939, 9, 1167, 1176. " Id., Z, anorg, Chem. 1936, 237, 81.

JO Group 0. Inert Gases gaseous phase. He also65 prepared with phenol the solid Xe, 2C6H5OH, which had a dissociation pressure of xenon of 1 atmosphere at 0 C. (the v.p. of liquid xenon at 0 C. is about 46 atmospheres).

Booth and Willson66 measured the freezing-point curve for the system A-BF3, and got maxima for the compositions 1 argon to 1, 2, 3, 6, 8, and 10 BF3; pressure up to 40 atmospheres had no effect. I t is very singular that all the 6 maxima and the melting-point of BF3 lie between 127 and 129 C.; the melting-points of the components are argon 189*4, BF3 127 C. At higher temperatures (mostly from 40 to 0 C.) they find no evidence of the existence of compounds.67

We must conclude that apart from the molecular ions occurring in the gas, there is in no case satisfactory evidence of the existence of chemical compounds of any of the inert gases.

65 B. A. Nikitin, CE. Acad. Sd. U.S.S.R. 1940, 29 , 571.

66 H. S. Booth and K. S. Willson, J.A.C.S. 1935, 57, 2273.

67 Ib . 2280.

GROUP I HYDROGEN, DEUTERIUM, AND TRITIUM

HYDROGEN, the lightest of the elements, was one of the chief problems of the original Periodic Table, since it has close affinities both with the alkali metals of Group I and with the halogens of Oioup VII. We now realize that it stands at the head of both groups, resembling the alkali metals in having a single easily detached electron, and the halogens in having one electron less than the next following inert gas. I t thus occupies a unique position in the Table.

The amount of hydrogen in the earth's crust, including the water and the air, is estimated at 0*87 per cent, by weight and 15*4 per cent, by atoms. Free hydrogen occurs to a minute extent in the atmosphere {according to Paneth1 less than 1 part in a million by volume); it is con-tinually being produced on the earth's surface from various sources, in-cluding oil gas outflows, but at the same time the molecules move quick enough to escape from the earth's gravitational field (see above, p. 2). Natural gas may contain up to 10 or even 30 per cent, of hydrogen, the remainder being mainly methane and ethane. Commercially hydrogen can be obtained from this source or from coal gas (of which it forms some 50 per cent.) by liquefying the other components, but it is more often got (for example in making synthetic ammonia) from water gas, which is a mixture of hydrogen and carbon monoxide, made by passing steam over heated coal or coke; if this is mixed with more steam and passed over it suitable catalyst (oxides of iron and cobalt are commonly used) at a temperature not above 400 C, the carbon monoxide reacts with the steam to give carbon dioxide and hydrogen. The carbon dioxide is removed by washing with water under pressure, and the residual carbon monoxide by treatment with ammoniacal cuprous solution, or by passing over heated noda lime, which reacts with it to give sodium formate.

Hydrogen occurs in a surprisingly large number of forms.

I. There are three known isotopes of hydrogen, of mass-numbers 1, 2, And 3. The first two of these occur in nature in the proportions roughly Of 0,000 to 1; the third is now known not to occur in natural hydrogen In detectable amounts, but it can be made by atomic bombardments, for iiutmple, of deuterium by deuterons. These isotopes are distinguished as tProtium, Deuterium, and Tritium; but the properties of pure protium and lt compounds are practically identical with those of the natural isotopic mixture. The peculiar properties of deuterium, and so far as they are known of tritium, are described in the next sections (deuterium, p. 36; tritium, p. 57).

1 F, A, Paneih, Nature, 1937, 139, 181.

12 Group / . Hydrogen II. Elementary hydrogen exists in several states. 1. Diatomic hydrogen H2, the normal state of the gas. This can assume

two forms with different properties, known as ortho- and para-hydrogen, and is normally a mixture of the two; the same is true of deuterium D2. The molecule can also have a positive charge H2^, as in positive ray tubes.

2. Monatomic or 'active' hydrogen H (see p. 15). This electrically neutral form can be produced in the gas by electrical excitement, and shows an intense chemical activity. The positively charged hydrogen atom H+ is the hydrogen nucleus, proton, or hydrogen ion (formerly thought to be the unit of positive electricity), which is important both from its chemical activity, and as a projectile for, and a product of, nuclear disintegration.

3. Triatomic hydrogen H3 certainly occurs as the positive ion H34" in positive ray tubes; its existence as a neutral H3 molecule, though it has often been asserted, is very doubtful.

III. In its compounds hydrogen is found: 1. As a positive ion H+ (except in the gas always solvated, as H+-X). 2. As a negative ion H" (with two unshared electrons), only in the

hydrides of the alkalis and the alkaline earths, such as Li[H] and Ca[H]2. 3. In the covalent form HX (with two shared electrons), as in the

hydrocarbons, and organic compounds generally. 4. In the form of the 'hydrogen bond' or (perhaps better) the 'hydrogen

bridge' H * , as in the ion [FH F]" or in associated hydroxylic compounds ROH * 0

Ortho- and Para-Hydrogen 13 The electrons must spin in opposite directions (antiparallel), or the mole-cule would not hold together; but the spins of the nuclei may either be in the same (parallel) or in opposite directions, and certain properties of the molecule, especially its specific heat, will be different in the two arrangements. In 1927 Dennison pointed out4 that the observed specific heats could be made to agree with the theory only if it was assumed that the time of transition was very long in comparison with the time in which the specific heats were measured, so that the gas was in effect a mixture of the two forms in fixed proportions. Subsequent work has entirely con-firmed this view. The two differ especially in the fact that of the rotational energy (which is practically all the heat energy other than translational that the molecules have up to high temperatures) the ortho (parallel) molecules have the odd quantum states 1, 3, 5, etc., and the para (anti-parallel) the even, 0, 2, 4, etc., which results in the ortho having, especially at low temperatures, a smaller specific heat than the para. We thus have:

Ortho Para

(Oj (OJ (OJ ^) Spins parallel Spins antiparallel Rotational quanta odd Rotational quanta even

The proportions can be determined from the heat conductivity (which depends on the specific heat) by observing the rate of cooling of a heated wire in the gas,

At equilibrium at the ordinary temperature and above it the gas con-tains 25 per cent, of para and 75 per cent, of ortho, but on cooling the percentage of para rises, because more of the gas can thus go into the (even) zero-quantum state, and at the boiling-point of hydrogen (20*4 K.) only a fraction of 1 per cent, of ortho remains. The proportions at equi-librium and the specific heats are:

Temp. Per cent, para Hp. ht. Pure para

Pure ortho Normal* H2

2O0K. 99-8

50 K. 76-9 0040 0000 0010

100 K. 38-5

1-504 0-073 0-431

200 K. 26-0 .

2-767 1151 1-555

298 K. 251 2-186 1-838 1-925

Inf. 2500

* i.e. the equilibrium mixture at the ordinary temperature, which has 25 per umit. para.

In order to make para-hydrogen (it is of course impossible to get the ortho with less than 25 per cent, of para) the gas must be cooled (if possible with liquid hydrogen, though liquid air under reduced pressure will give a gas with 45 per cent, of para), and then brought to equilibrium by means of a catalyst. This is most easily done by absorbing the gas on charcoal, and then after some minutes or hours, according to the activity of the the charcoal, pumping it off.

* D. M. Dennison, Proo. Boy, Soo, 1927 115, 483,

14 Group I. Hydrogen The physical properties differ slightly, &nd the values for some of them

(those for pure ortho being got by extrapolation) are2*5'6:

Pure para Pure ortho Normal (25 per cent, p.) .

M.-pL

13-88 K. 13-93 13-92

B.-pt.

20-29 20-41 20-38

MoL vol 200K. 28-54 28-35 28-40

Pure para-hydrogen remains unaltered in the gaseous state for weeks; the half-life is calculated to be about 3 years. The interconversion can be effected (1) by heat, at about 1,000 C.; (2) slowly in the liquid or solid state: the half-life in the liquid is about 5 days; (3) by treatment with atomic hydrogen (this is probably the mechanism of most of the con-versions): H + H 2 ( o ) - + H a ( P ) + H; (4) by paramagnetic molecules (including contact with paramagnetic metals, oxides, and salts7; and (5) by the catalytic action of certain surfaces.

The last three of these methods have led to interesting results, and are likely to lead to more. The conversion by atomic hydrogen (3) gives a method of measuring the concentration of free hydrogen atoms in a chemical reaction, and has been used for this purpose. In the para-magnetic method (4) it has been shown that the transformation is catalysed by oxygen, nitric oxide, and nitrogen peroxide, which are all paramagnetic, but not by such diamagnetic gases as nitrogen, nitrous oxide, carbon monoxide, or ammonia, nor by diborane B2H6, which therefore is pre-sumably diamagnetic also.

As to (5), a series of surfaces, especially platinum, and in a less degree nickel and copper (and to a small extent even sodium chloride), have been found to promote the change; this is no doubt due to the molecules breaking up, so that the hydrogen atoms may attach themselves separately to the surface.

This ortho-para difference should occur in all molecules containing two identical atoms if these atoms have a nuclear spin (in such molecules as He2, 12C2, 16O2, and 32S2 the nuclei have no spin). It may probably be found also with molecules of more than 2 atoms, such as H2O, but here we should expect the equilibrium always to be maintained. The only other molecule for which it has been established is deuterium D2, where the relations are much the same as with ordinary hydrogen, but the ratio of ortho to para at the ordinary temperature is 2:1 instead of 3:1.

For further information see Farkas8; for ortho- and para- D2 see below, p. 39,

6 K. Clusius and K. Hiller, Z. physihal. Chem. 1929, B 4, 158.

R. B. Scott and E. G. Briokwedde, J . Ohem. Phys. 1937, 5, 736. * H. S. Taylor and H. Diamond, J,A,0.8* 1835, 57, 1251. 8 A. F&fkai, Qrtho*hy$rog$n, PaM*hydfrog$n9 and Heavy Hydrogen, Cambridge,

1935.

Monatomic Hydrogen 15 The Molecule-ion HJ

This ion occurs in positive ray tubes,9 and also gives a characteristic band spectrum. It is of great theoretical interest, as the simplest possible case of two nuclei held together by a single electron: it contains the most certain 'one-electron bond'. The physics of this link was first worked out by Hund, a